Michael K. Y. Wong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Michael K. Y. Wong has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
Warren J. Cheung
Department of Emergency Medicine, University of Ottawa, Ottawa, Canada
3 shared publicationsCharisse Kwan
Department of Paediatrics, University of Western Ontario, London, Canada
2 shared publications- Christopher M. Yip
Institute of Biomedical Engineering, University of Toronto, Toronto, Canada
2 shared publications David Lewis
Saint John Regional Hospital, Dalhousie University, Saint John, Canada
2 shared publications- Dennis T. Ko
University of Toronto, Toronto, ON, Canada
2 shared publications - Harindra C. Wijeysundera
University of Toronto
2 shared publications
Latest Publications
- 4 March 2021
- Published by Springer Nature in Canadian Journal of Emergency Medicine
- Vol. 23 (4) , 450-454
Position statement: minimum archiving requirements for emergency medicine point-of-care ultrasound—a modified Delphi-derived national consensus
- 4 March 2021
- Published by Springer Nature in Canadian Journal of Emergency Medicine
- Vol. 23 (4) , 450-454
Position statement: minimum archiving requirements for emergency medicine point-of-care ultrasound—a modified Delphi-derived national consensus
- Open Access
- 27 May 2020
- Published by Oxford University Press (OUP) in Age and Ageing
- Vol. 50 (1) , 242-247
Does the public feel prepared to be substitute decision-makers? Gaps in preparedness and support for a high school curriculum: a national survey
Background: Substitute decision-makers (SDMs) make decisions on behalf of patients who do not have capacity, in line with previously expressed wishes, values and beliefs. However, miscommunications and poor awareness of previous wishes often lead to inappropriate care. Increasing public preparedness to communicate on behalf of loved ones may improve care in patients requiring an SDM. Methods: We conducted an online survey in January 2019 with a representative sample of the Canadian population. The primary outcome was self-reported preparedness to be an SDM. The secondary outcome was support for a high school curriculum on the role of SDMs. The effect of socio-demographics, known enablers and barriers to acting as an SDM, and attitudes towards a high school curriculum were assessed using multivariate analysis. Results: Of 1,000 participants, 53.1% felt prepared to be an SDM, and 75.4% stated they understood their loved one’s values. However, only 55.6% reported having had a meaningful conversation with their loved one about values and wishes, and only 61.7% reported understanding the SDM role. Engagement in advance care planning for oneself was low (23.1%). Age, experience, training and comfort with communication were associated with preparedness in our multivariate analysis. A high school curriculum was supported by 61.1% of respondents, with 28.3% neutral and 10.6% against it. Interpretation: There is a gap between perceived and actual preparedness to be an SDM. Many report understanding their loved one’s values yet have not asked them about wishes in illness or end of life. The majority of respondents support high school education to improve preparedness.
- 27 May 2020
- Published by Oxford University Press (OUP) in Age and Ageing
- Vol. 50 (1) , 242-247
Open AccessDoes the public feel prepared to be substitute decision-makers? Gaps in preparedness and support for a high school curriculum: a national survey
Background: Substitute decision-makers (SDMs) make decisions on behalf of patients who do not have capacity, in line with previously expressed wishes, values and beliefs. However, miscommunications and poor awareness of previous wishes often lead to inappropriate care. Increasing public preparedness to communicate on behalf of loved ones may improve care in patients requiring an SDM. Methods: We conducted an online survey in January 2019 with a representative sample of the Canadian population. The primary outcome was self-reported preparedness to be an SDM. The secondary outcome was support for a high school curriculum on the role of SDMs. The effect of socio-demographics, known enablers and barriers to acting as an SDM, and attitudes towards a high school curriculum were assessed using multivariate analysis. Results: Of 1,000 participants, 53.1% felt prepared to be an SDM, and 75.4% stated they understood their loved one’s values. However, only 55.6% reported having had a meaningful conversation with their loved one about values and wishes, and only 61.7% reported understanding the SDM role. Engagement in advance care planning for oneself was low (23.1%). Age, experience, training and comfort with communication were associated with preparedness in our multivariate analysis. A high school curriculum was supported by 61.1% of respondents, with 28.3% neutral and 10.6% against it. Interpretation: There is a gap between perceived and actual preparedness to be an SDM. Many report understanding their loved one’s values yet have not asked them about wishes in illness or end of life. The majority of respondents support high school education to improve preparedness.
- Open Access
- 1 July 2019
- Published by CMA Impact Inc. in CMAJ Open
- Vol. 7 (3) , E573-E581
Gaps in public preparedness to be a substitute decision-maker and the acceptability of high school education on resuscitation and end-of-life care: a mixed-methods study
Background: When a patient is incapable of making medical decisions for him- or herself, a substitute decision-maker makes choices according to the patient’s previously expressed wishes, values and beliefs; however, little is known about public readiness to act as a substitute decision-maker in Canada. Our primary objective was to measure public self-reported preparedness to act as a substitute decision-maker, and explore the attitudes, barriers and enablers associated with preparedness. Methods: From November 2017 to June 2018, we conducted a mixed-methods street intercept survey at 12 pedestrian areas in Ottawa, Ontario. We used descriptive statistics and logistic regression analysis to assess predictors of perceived preparedness to be a substitute decision-maker and determine support for high school education. We analyzed qualitative interview questions using inductive thematic analysis. Results: Of the 626 eligible respondents, 196 refused to participate, leaving 430 participants (response rate 68.7%). A total of 404 surveys (94.0%) were fully complete with no missing data. The respondents were mostly female (243 [56.5%]) and residents of Ontario (364 [84.6%]). The average age was 33.9 years. Although 314 respondents (73.0%) felt prepared to be a substitute decision-maker, 194 (45.1%) reported never having had meaningful conversations with loved ones to understand their wishes in the event of critical illness. A total of 293 participants (68.1%) identified important barriers to feeling prepared. Most respondents (309 [71.9%]) agreed that high school students should learn about being a substitute decision-maker, citing age appropriateness, potential societal benefit and improved decision-making, while cautioning the need to respect different maturity levels, cultures and experiences. Interpretation: he lack of conversation between loved ones reveals a gap between perceived and actual preparedness to be a substitute decision-maker for a loved one with a critical illness. The overall acceptability of high school education warrants further exploration.
- 1 July 2019
- Published by CMA Impact Inc. in CMAJ Open
- Vol. 7 (3) , E573-E581
Open AccessGaps in public preparedness to be a substitute decision-maker and the acceptability of high school education on resuscitation and end-of-life care: a mixed-methods study
Background: When a patient is incapable of making medical decisions for him- or herself, a substitute decision-maker makes choices according to the patient’s previously expressed wishes, values and beliefs; however, little is known about public readiness to act as a substitute decision-maker in Canada. Our primary objective was to measure public self-reported preparedness to act as a substitute decision-maker, and explore the attitudes, barriers and enablers associated with preparedness. Methods: From November 2017 to June 2018, we conducted a mixed-methods street intercept survey at 12 pedestrian areas in Ottawa, Ontario. We used descriptive statistics and logistic regression analysis to assess predictors of perceived preparedness to be a substitute decision-maker and determine support for high school education. We analyzed qualitative interview questions using inductive thematic analysis. Results: Of the 626 eligible respondents, 196 refused to participate, leaving 430 participants (response rate 68.7%). A total of 404 surveys (94.0%) were fully complete with no missing data. The respondents were mostly female (243 [56.5%]) and residents of Ontario (364 [84.6%]). The average age was 33.9 years. Although 314 respondents (73.0%) felt prepared to be a substitute decision-maker, 194 (45.1%) reported never having had meaningful conversations with loved ones to understand their wishes in the event of critical illness. A total of 293 participants (68.1%) identified important barriers to feeling prepared. Most respondents (309 [71.9%]) agreed that high school students should learn about being a substitute decision-maker, citing age appropriateness, potential societal benefit and improved decision-making, while cautioning the need to respect different maturity levels, cultures and experiences. Interpretation: he lack of conversation between loved ones reveals a gap between perceived and actual preparedness to be a substitute decision-maker for a loved one with a critical illness. The overall acceptability of high school education warrants further exploration.
- Open Access
- 2 May 2019
- Published by Springer Nature in Canadian Journal of Emergency Medicine
- Vol. 21 (S1) , S115-S116
P142: Gaps in public preparedness to be a substitute decision maker and the acceptability of high school education on resuscitation and end-of-life care: a mixed-methods study
Introduction: When a patient is incapable of making medical decisions for themselves, choices are made according to the patient's previously expressed, wishes, values, and beliefs by a substitute decision maker (SDM). While interventions to engage patients in their own advance care planning exist, little is known about public readiness to act as a SDM on behalf of a loved one. This mixed-methods survey aimed to describe attitudes, enablers and barriers to preparedness to act as a SDM, and support for a population-level curriculum on the role of an SDM in end-of-life and resuscitative care. Methods: From November 2017 to June 2018, a mixed-methods street intercept survey was conducted in Ottawa, Canada. Descriptive statistics and logistic regression analysis were used to assess predictors of preparedness to be a SDM and understand support for a high school curriculum. Responses to open-ended questions were analyzed using inductive thematic analysis. Results: The 430 respondents were mostly female (56.5%) with an average age of 33.9. Although 73.0% of respondents felt prepared to be a SDM, 41.0% of those who reported preparedness never had a meaningful conversation with loved ones about their wishes in critical illness. The only predictors of SDM preparedness were the belief that one would be a future SDM (OR 2.36 95% CI 1.34-4.17), and age 50-64 compared to age 16-17 (OR 7.46 95% CI 1.25-44.51). Thematic enablers of preparedness included an understanding of a patient's wishes, the role of the SDM and strong familial relationships. Barriers included cultural norms, family conflict, and a need for time for high stakes decisions. Most respondents (71.9%) believed that 16 year olds should learn about SDMs. They noted age appropriateness, potential developmental and societal benefit, and improved decision making, while cautioning the need for a nuanced approach respectful of different maturity levels, cultures and individual experiences. Conclusion: This study reveals a concerning gap between perceived preparedness and actions taken in preparation to be an SDM for loved ones suffering critical illness. The results also highlight the potential role for high school education to address this gap. Future studies should further explore the themes identified to inform development of resources and curricula for improved health literacy in resuscitation and end-of-life care.
- 2 May 2019
- Published by Springer Nature in Canadian Journal of Emergency Medicine
- Vol. 21 (S1) , S115-S116
Open AccessP142: Gaps in public preparedness to be a substitute decision maker and the acceptability of high school education on resuscitation and end-of-life care: a mixed-methods study
Introduction: When a patient is incapable of making medical decisions for themselves, choices are made according to the patient's previously expressed, wishes, values, and beliefs by a substitute decision maker (SDM). While interventions to engage patients in their own advance care planning exist, little is known about public readiness to act as a SDM on behalf of a loved one. This mixed-methods survey aimed to describe attitudes, enablers and barriers to preparedness to act as a SDM, and support for a population-level curriculum on the role of an SDM in end-of-life and resuscitative care. Methods: From November 2017 to June 2018, a mixed-methods street intercept survey was conducted in Ottawa, Canada. Descriptive statistics and logistic regression analysis were used to assess predictors of preparedness to be a SDM and understand support for a high school curriculum. Responses to open-ended questions were analyzed using inductive thematic analysis. Results: The 430 respondents were mostly female (56.5%) with an average age of 33.9. Although 73.0% of respondents felt prepared to be a SDM, 41.0% of those who reported preparedness never had a meaningful conversation with loved ones about their wishes in critical illness. The only predictors of SDM preparedness were the belief that one would be a future SDM (OR 2.36 95% CI 1.34-4.17), and age 50-64 compared to age 16-17 (OR 7.46 95% CI 1.25-44.51). Thematic enablers of preparedness included an understanding of a patient's wishes, the role of the SDM and strong familial relationships. Barriers included cultural norms, family conflict, and a need for time for high stakes decisions. Most respondents (71.9%) believed that 16 year olds should learn about SDMs. They noted age appropriateness, potential developmental and societal benefit, and improved decision making, while cautioning the need for a nuanced approach respectful of different maturity levels, cultures and individual experiences. Conclusion: This study reveals a concerning gap between perceived preparedness and actions taken in preparation to be an SDM for loved ones suffering critical illness. The results also highlight the potential role for high school education to address this gap. Future studies should further explore the themes identified to inform development of resources and curricula for improved health literacy in resuscitation and end-of-life care.
- 1 August 2015
- Published by Elsevier in Journal of Molecular Biology
- Vol. 427 (16) , 2599-2609
Mitochondrial Genome Maintenance 1 (Mgm1) Protein Alters Membrane Topology and Promotes Local Membrane Bending
- 1 August 2015
- Published by Elsevier in Journal of Molecular Biology
- Vol. 427 (16) , 2599-2609
Mitochondrial Genome Maintenance 1 (Mgm1) Protein Alters Membrane Topology and Promotes Local Membrane Bending
- Open Access
- 23 February 2015
- Published by CMA Impact Inc. in Canadian Medical Association Journal
- Vol. 187 (5) , E160-E168
Factors associated with physician follow-up among patients with chest pain discharged from the emergency department
Background: Many patients with chest pain do not receive follow-up from a physician after discharge from the emergency department despite significant survival benefit associated with follow-up care. Our objective was to evaluate factors associated with physician follow-up to understand this gap in practice. Methods: We conducted an observational study involving patients at high risk who were assessed for chest pain and discharged from an emergency department in Ontario between April 2004 and March 2010. We used multivariable logistic regression to determine the association of clinical and nonclinical characteristics with physician follow-up. Results: We identified 56 767 patients, of whom 25.1% did not receive any follow-up by a physician, 69.0% were seen by their primary care physician, and 17.3% were seen by a cardiologist within 30 days. Patients who had medical comorbidities and cardiac conditions such as myocardial infarction or heart failure were less likely to have follow-up. In contrast, a previous visit to a primary care physician was associated with the highest odds of having physician follow-up (odds ratio [OR] 6.44, 95% confidence interval [CI] 5.91–7.01). Similarly, a previous visit to a cardiologist was strongly associated with follow-up by a cardiologist (OR 3.01, 95% CI 2.85–3.17). Patients evaluated in emergency departments with the highest tertile of chest pain volume were more likely to receive follow-up from any physician (OR 1.52, 95% CI 1.31–1.77) and from a cardiologist (OR 2.04, 95% CI 1.61–2.57). Interpretation: Nonclinical factors are strongly associated with physician follow-up for patients with chest pain after discharge from the emergency department. However, patients with comorbidities and at higher risk for future adverse events are less likely to receive follow-up care.
- 23 February 2015
- Published by CMA Impact Inc. in Canadian Medical Association Journal
- Vol. 187 (5) , E160-E168
Open AccessFactors associated with physician follow-up among patients with chest pain discharged from the emergency department
Background: Many patients with chest pain do not receive follow-up from a physician after discharge from the emergency department despite significant survival benefit associated with follow-up care. Our objective was to evaluate factors associated with physician follow-up to understand this gap in practice. Methods: We conducted an observational study involving patients at high risk who were assessed for chest pain and discharged from an emergency department in Ontario between April 2004 and March 2010. We used multivariable logistic regression to determine the association of clinical and nonclinical characteristics with physician follow-up. Results: We identified 56 767 patients, of whom 25.1% did not receive any follow-up by a physician, 69.0% were seen by their primary care physician, and 17.3% were seen by a cardiologist within 30 days. Patients who had medical comorbidities and cardiac conditions such as myocardial infarction or heart failure were less likely to have follow-up. In contrast, a previous visit to a primary care physician was associated with the highest odds of having physician follow-up (odds ratio [OR] 6.44, 95% confidence interval [CI] 5.91–7.01). Similarly, a previous visit to a cardiologist was strongly associated with follow-up by a cardiologist (OR 3.01, 95% CI 2.85–3.17). Patients evaluated in emergency departments with the highest tertile of chest pain volume were more likely to receive follow-up from any physician (OR 1.52, 95% CI 1.31–1.77) and from a cardiologist (OR 2.04, 95% CI 1.61–2.57). Interpretation: Nonclinical factors are strongly associated with physician follow-up for patients with chest pain after discharge from the emergency department. However, patients with comorbidities and at higher risk for future adverse events are less likely to receive follow-up care.
- 18 November 2014
- Published by Wolters Kluwer Health in Circulation
- Vol. 130 (21) , 1883-1890
Trends in Short- and Long-Term Survival Among Out-of-Hospital Cardiac Arrest Patients Alive at Hospital Arrival
Background—: Out-of-hospital cardiac arrest (OHCA) is associated with a poor prognosis and poses a significant burden to the healthcare system, but few studies have evaluated whether OHCA incidence and survival have changed over time. Methods and Results—: A population-based cohort study was conducted, including 34 291 OHCA patients >20 years of age who were transported alive to the emergency department of an acute-care hospital from April 1, 2002, to March 31, 2012, in Ontario, Canada. Patients with life-threatening trauma and those who died before hospital arrival were excluded. The overall age- and sex-standardized incidence of OHCA patients who were transported alive was 36 cases per 100 000 persons and did not significantly change over the study period. Cardiac risk factor prevalence increased significantly, whereas the rate of most cardiovascular conditions decreased significantly. The 30-day survival improved from 9.4% in 2002 to 13.6% in 2011; 1-year survival improved from 7.7% to 11.8% ( P <0.001). Patients hospitalized in 2011 were significantly more likely to survive 30 days (adjusted odds ratio, 1.47 [95% CI, 1.22–1.77]) and 1 year (adjusted odds ratio, 1.55 [95% CI, 1.27–1.91]) compared with 2002. A significant interaction between temporal trends in survival improvement and age group was observed in which the improvement in survival was largest in the youngest age groups. Conclusions—: OHCA patients who were transported alive are increasingly likely to have cardiovascular risk factors but less likely to have previous cardiovascular conditions. The overall incidence of OHCA patients transported to hospital alive did not change over the past decade. Short- and longer-term survival after OHCA has substantially improved, with younger patients experiencing the greatest improvement.
- 18 November 2014
- Published by Wolters Kluwer Health in Circulation
- Vol. 130 (21) , 1883-1890
Trends in Short- and Long-Term Survival Among Out-of-Hospital Cardiac Arrest Patients Alive at Hospital Arrival
Background—: Out-of-hospital cardiac arrest (OHCA) is associated with a poor prognosis and poses a significant burden to the healthcare system, but few studies have evaluated whether OHCA incidence and survival have changed over time. Methods and Results—: A population-based cohort study was conducted, including 34 291 OHCA patients >20 years of age who were transported alive to the emergency department of an acute-care hospital from April 1, 2002, to March 31, 2012, in Ontario, Canada. Patients with life-threatening trauma and those who died before hospital arrival were excluded. The overall age- and sex-standardized incidence of OHCA patients who were transported alive was 36 cases per 100 000 persons and did not significantly change over the study period. Cardiac risk factor prevalence increased significantly, whereas the rate of most cardiovascular conditions decreased significantly. The 30-day survival improved from 9.4% in 2002 to 13.6% in 2011; 1-year survival improved from 7.7% to 11.8% ( P <0.001). Patients hospitalized in 2011 were significantly more likely to survive 30 days (adjusted odds ratio, 1.47 [95% CI, 1.22–1.77]) and 1 year (adjusted odds ratio, 1.55 [95% CI, 1.27–1.91]) compared with 2002. A significant interaction between temporal trends in survival improvement and age group was observed in which the improvement in survival was largest in the youngest age groups. Conclusions—: OHCA patients who were transported alive are increasingly likely to have cardiovascular risk factors but less likely to have previous cardiovascular conditions. The overall incidence of OHCA patients transported to hospital alive did not change over the past decade. Short- and longer-term survival after OHCA has substantially improved, with younger patients experiencing the greatest improvement.
- 10 December 2013
- Published by SAGE Publications in Journal of Alzheimer’s Disease
- Vol. 38 (4) , 907-922
Risk Factors for Development of Dementia in a Unique Six-Year Cohort Study. I. An Exploratory, Pilot Study of Involvement of the E4 Allele of Apolipoprotein E, Mutations of the Hemochromatosis-HFE Gene, Type 2 Diabetes, and Stroke
Risk factors for dementia development are not well-defined. We evaluated several factors alone and in combination in a unique cohort of Caucasian volunteers over an approximately 6-year observation window using a nested case/control design. Factors included: apolipoprotein E (ApoE) gene variants (the E4 allele is the strongest confirmed genetic predisposing factor for Alzheimer's disease), the hemochromatosis-HFE gene mutations (H63D and C282Y), diabetes, and stroke. At study entry, subjects were ≥65 years of age (M ± SD = 73.0 ± 4.9), had an MMSE score ≥24, and no evidence of cerebrovascular disease or current depression. Genotyping was completed on 163 available DNA samples from three different groups at the study end: those who still had normal cognitive function; those who had developed dementia; and those with Mild Cognitive Impairment (MCI). Analyses were interpreted at the 95% confidence level without Bonferroni corrections. In the subgroup with dementia, all cases of diabetes were type 2 and present at study entry, whereas all strokes occurred during the study. The results highlight apparently synergistic interactions between genetic and medical risk factors for dementia development, gender differences in risk factors, and involvement of HFE mutations. Having E4 (i.e., either of E3/4 or E4/4), C282Y, H63D, diabetes, or stroke alone did not attain significance. Significant predisposing factors with post-hoc power ≥80% were: E4 homozygosity (E4/4)males+females, odds ratio (OR) = 56.0); E4+diabetes (males+females, OR = 13.7; E4+H63D+diabetes (females, OR = 52.0); E4+stroke (males, OR = 46.5). The importance of preventing diabetes and stroke to ward off dementia and the possible role of iron dysmetabolism in dementia are discussed.
- 10 December 2013
- Published by SAGE Publications in Journal of Alzheimer’s Disease
- Vol. 38 (4) , 907-922
Risk Factors for Development of Dementia in a Unique Six-Year Cohort Study. I. An Exploratory, Pilot Study of Involvement of the E4 Allele of Apolipoprotein E, Mutations of the Hemochromatosis-HFE Gene, Type 2 Diabetes, and Stroke
Risk factors for dementia development are not well-defined. We evaluated several factors alone and in combination in a unique cohort of Caucasian volunteers over an approximately 6-year observation window using a nested case/control design. Factors included: apolipoprotein E (ApoE) gene variants (the E4 allele is the strongest confirmed genetic predisposing factor for Alzheimer's disease), the hemochromatosis-HFE gene mutations (H63D and C282Y), diabetes, and stroke. At study entry, subjects were ≥65 years of age (M ± SD = 73.0 ± 4.9), had an MMSE score ≥24, and no evidence of cerebrovascular disease or current depression. Genotyping was completed on 163 available DNA samples from three different groups at the study end: those who still had normal cognitive function; those who had developed dementia; and those with Mild Cognitive Impairment (MCI). Analyses were interpreted at the 95% confidence level without Bonferroni corrections. In the subgroup with dementia, all cases of diabetes were type 2 and present at study entry, whereas all strokes occurred during the study. The results highlight apparently synergistic interactions between genetic and medical risk factors for dementia development, gender differences in risk factors, and involvement of HFE mutations. Having E4 (i.e., either of E3/4 or E4/4), C282Y, H63D, diabetes, or stroke alone did not attain significance. Significant predisposing factors with post-hoc power ≥80% were: E4 homozygosity (E4/4)males+females, odds ratio (OR) = 56.0); E4+diabetes (males+females, OR = 13.7; E4+H63D+diabetes (females, OR = 52.0); E4+stroke (males, OR = 46.5). The importance of preventing diabetes and stroke to ward off dementia and the possible role of iron dysmetabolism in dementia are discussed.
- 21 November 2011
- Published by IOP Publishing in Nanotechnology
- Vol. 22 (49) , 494018
Nucleation and growth of elastin-like peptide fibril multilayers: anin situatomic force microscopy study
Controlling how molecules assemble into complex supramolecular architectures requires careful consideration of the subtle inter- and intra-molecular interactions that control their association. This is particularly crucial in the context of assembly at interfaces, where both surface chemistry and structure can play a role in directing structure formation. We report here the results of a study into the self-assembly of the elastin-like peptide EP I on structurally modified highly ordered pyrolytic graphite, including the role of spatial confinement on fibril nucleation and the growth of oriented fibril multilayers. In situ atomic force microscopy performed in fluid and at elevated temperature provided direct evidence of frustrated fibril nuclei and oriented growth of independent fibril domains. These results portend the application of this in situ strategy for studies of the nucleation and growth mechanisms of other fibril- and amyloid-forming proteins.
- 21 November 2011
- Published by IOP Publishing in Nanotechnology
- Vol. 22 (49) , 494018
Nucleation and growth of elastin-like peptide fibril multilayers: anin situatomic force microscopy study
Controlling how molecules assemble into complex supramolecular architectures requires careful consideration of the subtle inter- and intra-molecular interactions that control their association. This is particularly crucial in the context of assembly at interfaces, where both surface chemistry and structure can play a role in directing structure formation. We report here the results of a study into the self-assembly of the elastin-like peptide EP I on structurally modified highly ordered pyrolytic graphite, including the role of spatial confinement on fibril nucleation and the growth of oriented fibril multilayers. In situ atomic force microscopy performed in fluid and at elevated temperature provided direct evidence of frustrated fibril nuclei and oriented growth of independent fibril domains. These results portend the application of this in situ strategy for studies of the nucleation and growth mechanisms of other fibril- and amyloid-forming proteins.